pith. sign in

arxiv: 2411.01799 · v4 · submitted 2024-11-04 · 💰 econ.EM

Estimating Nonseparable Selection Models: A Functional Contraction Approach

Fan Wu , Yi Xin This is my paper

Pith reviewed 2026-05-23 18:04 UTC · model grok-4.3

classification 💰 econ.EM
keywords nonseparable selection modelscontraction mappingnonparametric identificationtwo-step estimationpotential outcomesselection biaseconometrics
0
0 comments X

The pith

For a given selection function, potential outcome distributions in nonseparable models are nonparametrically identified from selected outcomes via contraction mapping iteration.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper establishes that, given a selection function, the distributions of potential outcomes in nonseparable selection models can be nonparametrically identified from the distributions of selected outcomes. This identification is achieved through a simple iterative procedure that relies on a contraction mapping property. The approach permits full-information estimation of the model without requiring parametric assumptions or separability between the outcome and selection equations. As a result, it supports consistent estimation in a variety of empirical contexts such as demand systems, auctions, and labor market models.

Core claim

For a given selection function, the potential outcome distributions are nonparametrically identified from the selected outcome distributions and can be recovered using a simple iterative algorithm based on a contraction mapping. This result enables a full-information approach to estimating selection models without imposing parametric or separability assumptions on the outcome equation. The authors propose a two-step estimation strategy for the potential outcome distributions and the parameters of the selection function and establish the consistency and asymptotic normality of the resulting estimators.

What carries the argument

A contraction mapping that iteratively recovers potential outcome distributions from observed selected outcome distributions.

If this is right

  • The resulting estimators for potential outcome distributions and selection function parameters are consistent and asymptotically normal.
  • The method applies directly to settings with only selected outcomes observed, such as consumer demand models using transaction prices alone.
  • It covers auctions with incomplete bid data and Roy models with data only on accepted wages.
  • Monte Carlo simulations confirm reliable finite-sample performance across these applications.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The contraction approach could be nested inside a joint estimation routine that treats the selection function as unknown but parameterized, allowing one-step rather than two-step inference.
  • Varying the assumed selection function over a range and tracing the implied potential distributions could yield partial identification bounds for counterfactuals.
  • The same mapping principle might extend to selection problems in other fields, such as recovering full distributions from censored medical trial data.
  • Testing whether the estimated distributions satisfy separability restrictions could serve as a specification check against conventional estimators.

Load-bearing premise

The selection function is treated as given or separately parameterized, and the contraction mapping has a unique fixed point that recovers the true potential outcome distributions.

What would settle it

Generating data from a known nonseparable selection model and checking whether the iterative algorithm converges to the true potential distributions rather than the selected ones would test the identification claim.

Figures

Figures reproduced from arXiv: 2411.01799 by Fan Wu, Yi Xin.

Figure 1
Figure 1. Figure 1: CDF of log(price) for firms 1 and 2 (conditional on [PITH_FULL_IMAGE:figures/full_fig_p028_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Estimated density functions price by exploiting a one-to-one mapping between an order statistic and its parent distribution. Komarova (2013) analyzes asymmetric second-price auctions where only the winning bids and the winner’s identity are observed. A related result for general￾ized competing risks models can be found in Meilijson (1981). More recently, Guerre and Luo (2019) examine nonparametric identifi… view at source ↗
read the original abstract

We propose a novel method for estimating nonseparable selection models. We show that, for a given selection function, the potential outcome distributions are nonparametrically identified from the selected outcome distributions and can be recovered using a simple iterative algorithm based on a contraction mapping. This result enables a full-information approach to estimating selection models without imposing parametric or separability assumptions on the outcome equation. We propose a two-step estimation strategy for the potential outcome distributions and the parameters of the selection function and establish the consistency and asymptotic normality of the resulting estimators. Monte Carlo simulations demonstrate that our approach performs well in finite samples. The method is applicable to a wide range of empirical settings, including consumer demand models with only transaction prices, auctions with incomplete bid data, and Roy models with data on accepted wages.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 2 minor

Summary. The paper proposes a novel method for estimating nonseparable selection models. It shows that, for a given selection function, the potential outcome distributions are nonparametrically identified from the selected outcome distributions and can be recovered using a simple iterative algorithm based on a contraction mapping. This enables a two-step estimation strategy for the potential outcome distributions and the parameters of the selection function, with established consistency and asymptotic normality. Monte Carlo simulations demonstrate good finite-sample performance. The method applies to settings such as consumer demand with transaction prices, auctions with incomplete bids, and Roy models with accepted wages.

Significance. If the contraction-mapping identification result holds under the stated conditions, the paper provides a useful nonparametric route to identification and estimation in selection models without requiring separability or parametric restrictions on the outcome equation. The iterative recovery algorithm is computationally attractive, and the two-step estimator with asymptotic theory supports practical use. Monte Carlo evidence is a strength, though the absence of real-data illustrations limits immediate empirical assessment.

major comments (2)
  1. [Estimation and asymptotics sections] The identification argument is explicitly conditional on a fixed selection function, but the two-step procedure estimates its parameters; the manuscript should derive how first-step estimation error propagates into the second-step asymptotic normality and consistency rates (see the estimation and asymptotics sections).
  2. [Identification section] The contraction-mapping property is asserted to deliver unique nonparametric identification, but the specific function space, metric, and bound on the contraction constant (strictly less than one) must be verified explicitly for the selection functions used in the consumer-demand, auction, and Roy-model examples.
minor comments (2)
  1. [Abstract] The abstract states the identification result but does not list the precise conditions on the selection function required for the contraction to hold; adding a brief clause would improve clarity.
  2. [Monte Carlo section] Monte Carlo results would benefit from reported standard errors or confidence bands around the bias and RMSE figures, as well as explicit rules for any data trimming or support restrictions.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We appreciate the referee's thoughtful review and recommendation for minor revision. We address the major comments below and will incorporate the suggested clarifications in the revised manuscript.

read point-by-point responses

  1. Referee: [Estimation and asymptotics sections] The identification argument is explicitly conditional on a fixed selection function, but the two-step procedure estimates its parameters; the manuscript should derive how first-step estimation error propagates into the second-step asymptotic normality and consistency rates (see the estimation and asymptotics sections).

    Authors: We agree with this observation. The current manuscript establishes consistency and asymptotic normality under the assumption of a known selection function in the identification step, but the two-step estimator involves estimating the selection parameters first. In the revision, we will extend the asymptotic analysis to account for the first-step estimation error, deriving the appropriate influence functions and verifying that the rates remain valid for the second-step estimators under standard regularity conditions. revision: yes

  2. Referee: [Identification section] The contraction-mapping property is asserted to deliver unique nonparametric identification, but the specific function space, metric, and bound on the contraction constant (strictly less than one) must be verified explicitly for the selection functions used in the consumer-demand, auction, and Roy-model examples.

    Authors: We thank the referee for pointing this out. While the general contraction mapping theorem is applied in the identification section, we will add explicit verifications in an appendix or dedicated subsection for each empirical example. This will include specifying the function space (e.g., the space of bounded continuous functions equipped with the sup norm), the metric, and demonstrating that the contraction constant is strictly less than one for the relevant selection functions in the consumer demand, auction, and Roy model settings, using the maintained assumptions on the distributions and selection mechanisms. revision: yes

Circularity Check

0 steps flagged

No significant circularity; derivation self-contained

full rationale

The paper's central identification result is stated as nonparametric recovery of potential outcome distributions from selected distributions via an iterative contraction mapping that is explicitly conditional on a fixed (given or separately estimated) selection function. The subsequent two-step estimator for selection parameters is described as standard and follows the mapping step. No equations reduce the target distributions to fitted inputs by construction, no self-citations are invoked as load-bearing uniqueness theorems, and no ansatz or renaming of known results is presented as a derivation. The approach is therefore independent of the quantities it claims to recover.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

Review based solely on abstract; full paper would be needed to enumerate all modeling assumptions.

axioms (1)
  • domain assumption The selection model admits a contraction mapping that uniquely recovers the potential outcome distributions from the selected distributions.
    This property is invoked as the basis for nonparametric identification in the abstract.

pith-pipeline@v0.9.0 · 5652 in / 1118 out tokens · 27256 ms · 2026-05-23T18:04:04.353946+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

68 extracted references · 68 canonical work pages

  1. [1]

    and Powell, J

    Ahn, H. and Powell, J. L. (1993). Semiparametric estimation of censored selection models with a nonparametric selection mechanism. Journal of Econometrics , 58(1-2):3--29

    work page 1993

  2. [2]

    Allen, J., Clark, R., Hickman, B., and Richert, E. (2024). Resolving failed banks: Uncertainty, multiple bidding and auction design. Review of Economic Studies , 91(3):1201--1242

    work page 2024

  3. [3]

    Allen, J., Clark, R., and Houde, J.-F. (2014). Price dispersion in mortgage markets. The Journal of Industrial Economics , 62(3):377--416

    work page 2014

  4. [4]

    Allen, J., Clark, R., and Houde, J.-F. (2019). Search frictions and market power in negotiated-price markets. Journal of Political Economy , 127(4):1550--1598

    work page 2019

  5. [5]

    Andrews, D. W. and Schafgans, M. M. (1998). Semiparametric estimation of the intercept of a sample selection model. The Review of Economic Studies , 65(3):497--517

    work page 1998

  6. [6]

    and Bonhomme, S

    Arellano, M. and Bonhomme, S. (2017). Quantile selection models with an application to understanding changes in wage inequality. Econometrica , 85(1):1--28

    work page 2017

  7. [7]

    and Cantillon, E

    Asker, J. and Cantillon, E. (2008). Properties of scoring auctions. The RAND Journal of Economics , 39(1):69--85

    work page 2008

  8. [8]

    and Haile, P

    Athey, S. and Haile, P. A. (2002). Identification of standard auction models. Econometrica , 70(6):2107--2140

    work page 2002

  9. [9]

    and Haile, P

    Athey, S. and Haile, P. A. (2007). Nonparametric approaches to auctions. Handbook of econometrics , 6:3847--3965

    work page 2007

  10. [10]

    Baricz, \'A . (2008). Mills' ratio: Monotonicity patterns and functional inequalities. Journal of Mathematical Analysis and Applications , 340(2):1362--1370

    work page 2008

  11. [11]

    Bayer, P., Khan, S., and Timmins, C. (2011). Nonparametric identification and estimation in a roy model with common nonpecuniary returns. Journal of Business & Economic Statistics , 29(2):201--215

    work page 2011

  12. [12]

    Berry, S., Levinsohn, J., and Pakes, A. (1995). Automobile prices in market equilibrium. Econometrica , 63(4):841--890

    work page 1995

  13. [13]

    Berry, S. T. (1994). Estimating discrete-choice models of product differentiation. The RAND Journal of Economics , pages 242--262

    work page 1994

  14. [14]

    Borjas, G. (1987). Self-selection and the earnings of immigrants. American Economic Review , 77:531--553

    work page 1987

  15. [15]

    Buchholz, N., Doval, L., Kastl, J., Mat e jka, F., and Salz, T. (2020). The value of time: Evidence from auctioned cab rides. CEPR Discussion Paper No. DP14666

    work page 2020

  16. [16]

    Campo, S. (2012). Risk aversion and asymmetry in procurement auctions: Identification, estimation and application to construction procurements. Journal of Econometrics , 168(1):96--107

    work page 2012

  17. [17]

    A., Mogstad, M., and Mountjoy, J

    Canay, I. A., Mogstad, M., and Mountjoy, J. (2024). On the use of outcome tests for detecting bias in decision making. Review of Economic Studies , 91(4):2135--2167

    work page 2024

  18. [18]

    and Khan, S

    Chen, S. and Khan, S. (2003). Semiparametric estimation of a heteroskedastic sample selection model. Econometric Theory , 19(6):1040--1064

    work page 2003

  19. [19]

    Chernozhukov, V., Fern \'a ndez-Val, I., and Luo, S. (2023). Distribution regression with sample selection and uk wage decomposition. Technical report, cemmap working paper

    work page 2023

  20. [20]

    Cicala, S. (2015). When does regulation distort costs? lessons from fuel procurement in us electricity generation. American Economic Review , 105(1):411--444

    work page 2015

  21. [21]

    and Kitamura, Y

    Compiani, G. and Kitamura, Y. (2016). Using mixtures in econometric models: a brief review and some new results. The Econometrics Journal , 19(3):C95--C127

    work page 2016

  22. [22]

    Cosconati, M., Xin, Y., Wu, F., and Jin, Y. (2024). Competing under information heterogeneity: Evidence from auto insurance. Working paper

    work page 2024

  23. [23]

    S., Pavanini, N., and Schivardi, F

    Crawford, G. S., Pavanini, N., and Schivardi, F. (2018). Asymmetric information and imperfect competition in lending markets. American Economic Review , 108(7):1659--1701

    work page 2018

  24. [24]

    K., and Vella, F

    Das, M., Newey, W. K., and Vella, F. (2003). Nonparametric estimation of sample selection models. The Review of Economic Studies , 70(1):33--58

    work page 2003

  25. [25]

    G., Kosmopoulou, G., and Lamarche, C

    De Silva, D. G., Kosmopoulou, G., and Lamarche, C. (2009). The effect of information on the bidding and survival of entrants in procurement auctions. Journal of Public Economics , 93(1-2):56--72

    work page 2009

  26. [26]

    and Misra, S

    Dub \'e , J.-P. and Misra, S. (2023). Personalized pricing and consumer welfare. Journal of Political Economy , 131(1):131--189

    work page 2023

  27. [27]

    and Maurel, A

    d’Haultfoeuille, X. and Maurel, A. (2013a). Another look at the identification at infinity of sample selection models. Econometric Theory , 29(1):213--224

    work page

  28. [28]

    and Maurel, A

    d’Haultfoeuille, X. and Maurel, A. (2013b). Inference on an extended roy model, with an application to schooling decisions in france. Journal of Econometrics , 174(2):95--106

    work page

  29. [29]

    D’Haultf uille, X., Maurel, A., and Zhang, Y. (2018). Extremal quantile regressions for selection models and the black--white wage gap. Journal of Econometrics , 203(1):129--142

    work page 2018

  30. [30]

    Fernández-Val, I., van Vuuren , A., and Vella, F. (2024). Nonseparable sample selection models with censored selection rules. Journal of Econometrics , 240(2):105088

    work page 2024

  31. [31]

    and Perrigne, I

    Flambard, V. and Perrigne, I. (2006). Asymmetry in procurement auctions: Evidence from snow removal contracts. The Economic Journal , 116(514):1014--1036

    work page 2006

  32. [32]

    and Taber, C

    French, E. and Taber, C. (2011). Identification of models of the labor market. In Handbook of labor economics , volume 4, pages 537--617. Elsevier

    work page 2011

  33. [33]

    Goldberg, P. K. (1996). Dealer price discrimination in new car purchases: Evidence from the consumer expenditure survey. Journal of Political Economy , 104(3):622--654

    work page 1996

  34. [34]

    Gronau, R. (1974). Wage comparisons--a selectivity bias. Journal of political Economy , 82(6):1119--1143

    work page 1974

  35. [35]

    and Luo, Y

    Guerre, E. and Luo, Y. (2019). Nonparametric identification of first-price auction with unobserved competition: A density discontinuity framework. arXiv preprint arXiv:1908.05476

    work page arXiv 2019

  36. [36]

    Heckman, J. J. (1974). Shadow prices, market wages, and labor supply. Econometrica: journal of the econometric society , pages 679--694

    work page 1974

  37. [37]

    Heckman, J. J. (1976). The common structure of statistical models of truncation, sample selection and limited dependent variables and a simple estimator for such models. In Annals of economic and social measurement, volume 5, number 4 , pages 475--492. NBER

    work page 1976

  38. [38]

    Heckman, J. J. (1979). Sample selection bias as a specification error. Econometrica , 47:153--161

    work page 1979

  39. [39]

    Heckman, J. J. and Honore, B. E. (1990). The empirical content of the roy model. Econometrica: Journal of the Econometric Society , pages 1121--1149

    work page 1990

  40. [40]

    Heckman, J. J. and Sedlacek, G. (1985). Heterogeneity, aggregation, and market wage functions: An empirical model of self-selection in the labor market. Journal of political Economy , 93(6):1077--1125

    work page 1985

  41. [41]

    Heckman, J. J. and Vytlacil, E. J. (2007). Econometric evaluation of social programs, part i: Causal models, structural models and econometric policy evaluation. Handbook of econometrics , 6:4779--4874

    work page 2007

  42. [42]

    and Porter, R

    Hendricks, K. and Porter, R. H. (1988). An empirical study of an auction with asymmetric information. The American Economic Review , pages 865--883

    work page 1988

  43. [43]

    L., and Zhu, D

    Horta c su, A., Luco, F., Puller, S. L., and Zhu, D. (2019). Does strategic ability affect efficiency? evidence from electricity markets. American Economic Review , 109(12):4302--4342

    work page 2019

  44. [44]

    Hu, Y. (2017). The econometrics of unobservables--latent variable and measurement error models and their applications in empirical industrial organization and labor economics. Manuscript in preparation

    work page 2017

  45. [45]

    Ichimura, H. (1993). Semiparametric least squares (sls) and weighted sls estimation of single-index models. Journal of econometrics , 58(1-2):71--120

    work page 1993

  46. [46]

    Klein, R. W. and Spady, R. H. (1993). An efficient semiparametric estimator for binary response models. Econometrica: Journal of the Econometric Society , pages 387--421

    work page 1993

  47. [47]

    Komarova, T. (2013). A new approach to identifying generalized competing risks models with application to second-price auctions. Quantitative Economics , 4(2):269--328

    work page 2013

  48. [48]

    Krasnokutskaya, E., Song, K., and Tang, X. (2020). The role of quality in internet service markets. Journal of Political Economy , 128(1):75--117

    work page 2020

  49. [49]

    Lee, J. H. and Park, B. G. (2023). Nonparametric identification and estimation of the extended roy model. Journal of Econometrics , 235(2):1087--1113

    work page 2023

  50. [50]

    Lee, L.-F. (1978). Unionism and wage rates: A simultaneous equations model with qualitative and limited dependent variables. International economic review , pages 415--433

    work page 1978

  51. [51]

    Lee, L.-F. (1982). Some approaches to the correction of selectivity bias. The Review of Economic Studies , 49(3):355--372

    work page 1982

  52. [52]

    Lee, L.-F. (1983). Generalized econometric models with selectivity. Econometrica: Journal of the Econometric Society , pages 507--512

    work page 1983

  53. [53]

    and Bajari, P

    Lewis, G. and Bajari, P. (2011). Procurement contracting with time incentives: Theory and evidence. The Quarterly Journal of Economics , 126(3):1173--1211

    work page 2011

  54. [54]

    McKelvey, R. D. and Palfrey, T. R. (1995). Quantal response equilibria for normal form games. Games and Economic Behavior , 10(1):6--38

    work page 1995

  55. [55]

    Meilijson, I. (1981). Estimation of the lifetime distribution of the parts from the autopsy statistics of the machine. Journal of Applied Probability , 18(4):829--838

    work page 1981

  56. [56]

    Mourifie, I., Henry, M., and Meango, R. (2020). Sharp bounds and testability of a roy model of stem major choices. Journal of Political Economy , 128(8):3220--3283

    work page 2020

  57. [57]

    Nakabayashi, J. (2013). Small business set-asides in procurement auctions: An empirical analysis. Journal of Public Economics , 100:28--44

    work page 2013

  58. [58]

    Newey, W. K. (2007). Nonparametric continuous/discrete choice models. International Economic Review , 48(4):1429--1439

    work page 2007

  59. [59]

    Newey, W. K. (2009). Two-step series estimation of sample selection models. The Econometrics Journal , 12(suppl\_1):S217--S229

    work page 2009

  60. [60]

    Newey, W. K. and McFadden, D. (1994). Large sample estimation and hypothesis testing. Handbook of Econometrics , 4:2111--2245

    work page 1994

  61. [61]

    Roy, A. D. (1951). Some thoughts on the distribution of earnings. Oxford Economic Papers , 3(2):135--146

    work page 1951

  62. [62]

    Sagl, S. (2023). Dispersion, discrimination, and the price of your pickup. Working paper

    work page 2023

  63. [63]

    Salz, T. (2022). Intermediation and competition in search markets: An empirical case study. Journal of Political Economy , 130(2):310--345

    work page 2022

  64. [64]

    Takahashi, H. (2018). Strategic design under uncertain evaluations: Structural analysis of design-build auctions. The RAND Journal of Economics , 49(3):594--618

    work page 2018

  65. [65]

    Thompson, A. C. (1963). On certain contraction mappings in a partially ordered vector space. Proceedings of the American Mathematical Society , 14(3):438--443

    work page 1963

  66. [66]

    Vella, F. (1998). Estimating models with sample selection bias: a survey. Journal of Human Resources , pages 127--169

    work page 1998

  67. [67]

    Willis, R. J. and Rosen, S. (1979). Education and self-selection. Journal of Political Economy , 87(5, Part 2):S7--S36

    work page 1979

  68. [68]

    Yoganarasimhan, H. (2016). Estimation of beauty contest auctions. Marketing Science , 35(1):27--54

    work page 2016